Crop Life America President Jay Vroom was misquoted in a June 17 article on human pesticide tests. The word "effects" was substituted for "benefits" in the quotation, which should have read: "We think our products have enormous demonstrable benefits. They obviously have some risks." (Published 6/21/05)

The Environmental Protection Agency is using data from two dozen tests that deliberately exposed humans to toxic chemicals to help determine whether to approve new pesticides, according to a study released yesterday by congressional Democrats.

The lawmakers -- Sen. Barbara Boxer and Rep. Henry A. Waxman, both of California -- said the report underscored why EPA officials should no longer accept pesticide industry studies that involve human subjects. Volunteers were exposed to several poisons, including an insecticide used for chemical warfare in World War I and a pesticide closely related to the chemical that killed thousands in Bhopal, India.

Boxer, who plans to offer an amendment that would prohibit the EPA from using data culled from human pesticide tests, said in an interview it is "very disturbing" that officials "are in essence acquiescing to these kinds of experiments." She added, "The EPA has no standards in place, and they are in violation of international standards."

The EPA plans to issue new rules on human pesticide tests next year; in the interim, senior officials decide on a case-by-case basis whether the experiments violate ethical standards.

Policymakers have struggled for more than a decade over the propriety of using information from tests of pesticide effects on humans. The Clinton administration banned the use of such data in regulatory decisions, but Bush appointees revived the practice, relying on a 2004 National Academy of Sciences study that outlined human test guidelines. Some critics, however, say the EPA is not adhering to those guidelines.

Democratic staffers surveyed 24 tests conducted between 1967 and 2004, most in the past decade, that the EPA is reviewing as it evaluates applications to market new pesticides. They concluded that nearly one-third of the tests "were specifically designed to cause harm to human test subjects or put them at risk of harm," and in many cases "the informed consent forms used in the experiments do not appear to meet ethical standards."

The scientists conducting the tests frequently ignored the fact that they were putting their subjects -- who were often students or minorities -- at risk, the report said. In one case, three dozen subjects took an insecticide pill with orange juice at breakfast; in another, eight people received a dose of a toxic insecticide for 28 days, during which time all eight "experienced adverse events, including headaches, abdominal pain, nausea, coughing and a rash. The researchers declared that every single adverse event was unrelated to dosing."

EPA spokeswoman Eryn Witcher said the agency is drafting new testing rules. "The agency values the importance of the scientific and ethical issues surrounding human studies, which is why EPA is expediting the process to issue its first-ever regulation on third party studies," she said.

EPA Administrator Stephen L. Johnson has said human pesticide testing is not needed to protect public health, and earlier this year, under pressure from Democrats, he canceled an EPA study of how infants and small children are exposed to pesticides in their homes.

Last month, the House voted to bar the EPA from conducting or accepting any studies that test pesticides on humans. Some Senate Republicans, however, oppose such a provision and may fight Boxer's amendment.

Many pesticide industry officials see human tests as providing a clearer assessment of the environmental impact of their products than experiments with animals.

Jay Vroom, who represents major pesticide manufacturers as president of CropLife America, said using human clinical data "is somewhat unusual and fairly rare." Some of the tests Democrats focused on took place decades ago, he said, and current industry experiments comply with federal standards.

"We think our products have enormous demonstrable effects," Vroom said. "They obviously have some risks."